AMBER Archive (2003)

Subject: AMBER: problems with periodic boundary conditions

From: Monika Fuxreiter (monika_at_enzim.hu)
Date: Fri Sep 05 2003 - 04:17:40 CDT

    Hi,

     We have problems with running sander molecular dynamics with periodic
boundary conditions. The cell parameters are specified in the topology
file, and in the input file NTB=1 and NTP=0 are used. The topology file
is an old-style format, generated by amber 5.
 To run MD we are using sander in amber 7. What happens in the run,
that waters at the boundary start to move away and the density
condsiderably decreases near the edges. This does not seem to be a boxing
problem, that few waters are translated to other cells. It appears as a
problem with periodicity.
In the output, the program reports the cell parameters and also they are
appended to the rst file. In spite of this, the system behaves in a very
strange way. We have been using amber 5 many times, without having trouble
with the simulations.
 Can oneone tell me where the problem can come from? Can it be a
compilation problem, or the old-style topology or something else?

   Thanks

    Monika Fuxreiter

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu